Twitter algorithm changes have been accused of exacerbating anger and animosity among users, according to a recent study conducted by computer scientists at Cornell University and the University of California, Berkeley. The researchers examined tweets shown to 806 users in February, comparing the content displayed on Twitter’s personalized timelines and the chronological newsfeed. They discovered that the algorithm now emphasizes “emotional content,” leading users to react more strongly than in the past, even if the original intent of the tweets was not to provoke a response.
The study concluded that the algorithm amplifies tweets expressing strong emotions, particularly anger. Furthermore, exposure to algorithm-selected tweets resulted in increased emotional responses and a heightened sense of anger among users. Political tweets shown by the algorithm displayed greater partisanship and out-group animosity, potentially fueling negative beliefs about those with opposing views.
Interestingly, the algorithm slightly increased the ratio of out-group to in-group content, rather than reinforcing filter bubbles or echo chambers. This exposure to algorithm-selected content led users to perceive their political in-group more positively and the out-group more negatively, contributing to greater polarization.
The researchers also noted that users were more likely to follow recommended accounts and encounter similar emotional content on the platform. The study, conducted as a controlled experiment without internal access, emphasized the significance of understanding the impact of machine-learning algorithms on content curation and user behavior, particularly as social media continues to influence public opinion.
However, not everyone agrees with the study’s findings. Marvin Winkelmann, managing director of talent management and content creation marketing agency AFK, argues that contentious topics naturally generate more interaction and engagement. He believes that Twitter’s algorithm simply amplifies the visibility of controversial topics, leading to the perception of more negative comments.
Winkelmann suggests that the algorithm prioritizes contentious topics because they generate more interaction, regardless of whether the opinions expressed are positive or negative. He points out that even seemingly positive topics can spark disagreement and arguments in the comments section.
In conclusion, the study highlights the potential negative consequences of Twitter’s algorithm changes, particularly in terms of amplifying anger, animosity, and affective polarization. While some experts argue that contentious topics naturally drive engagement, the study’s findings suggest that the algorithm may be exacerbating these negative emotions. As social media’s influence on public opinion continues to grow, understanding the impact of algorithmic content curation remains crucial.